Course Number: Math 122

Descriptive Title: Advanced Statistics

Unit: 4 units

Contact Hours: 4hrs/wk.

Class Schedule: MW (1:00-3:00)

Math 121

•Definition/ Meaning of Statistics

Singular essence: Statistics is a branch of science that deals with the study of quantitative
data collection, presentation, analysis, interpretation

Plural essence: Statistics is a set or collection of quantitative data

•2 Division of Statistics

°Descriptive Statistics - used to describe, summarize, organize, present a dataset

°Inferential Statistics - used to formulate inference/conclusion/generalization from the

analysis of the sample

•Measures of Central Tendency

Scenario: The Municipality of Alangalang wants to be informed of how knowledgeable on

nCOV are households in urban and rural barangays in the municipality.

LO1: Describe and identify the concepts of universe, population, and variable in
research scenarios.

Universe - totality of objects under consideration

Example: All households in urban and rural barangays in the municipality of Alangalang,
Leyte

Variable - characteristic of any member of the universe that is measurable

Changes

- measurable characteristics

Example: knowledge

TWO TYPES OF VARIABLE

1. 1.Quantitative - direct measures

a. Continous - can be found in a particular continuum

- direct measurement

b. Discrete - frequency or counts

Example: number of students or head count

2. Qualitative -values are qualities

Ex. Sex (male and female)

It will only analyze if you will represent it in numbers. Male is 0 while female is 1.

4 Types of variables according to Level of measurements

1. Nominal - Does not have numerical meaning

- values are used to label or identify categories.

- MDAS is not allowed, count the males and females

Numbers are simply ideas/ theory.

2. Ordinal - Values have numerical meaning (rank or order)

- Example: Year classificaton of students
•Freshman = 1, Sophomore = 2, Junior = 3, Senior = 4
Count the number of students, sophomore...
- MDAS is not allowed
3. Interval - has no fixed point
- zero has diferent meanings
- Add and Subtract
°Scores
°Temperature (Celcius, Fahrenheit, Kelvin)
* Freezing point of water (Celcius = 0, Fahrenheit = 32, Kelvin =
273.15)
4. Ratio - has a fixed zero point
- Zero means absence of a measure
- Results of direct measurement
- MDAS

Number is just an Idea

Fully Knowledgeable - (80-100)

Partially Knowledgeable - (30-79)

Not Knowledgeable - (29 and below)

20 items - Perfect score of 20 (Raw Score) - Percent Score

Population - Set of all possible values of the variable

- Variable: Sex

- Scenario : The Division School Superintendent wants to be informed of the sex

distributipn of Grades 11 and 12 students in all public secondary schools in Deped Leyte
Division.

- Universe: Grades 11 and 12 students in all public secondary schools in Deped

Leyte Division.

Variable: Sex

TABLE 1. Sex distribution of Grades 11 and 12 students in all public secondary

schools in Deped Leyte Division.

Sex Frequency Relative Frquency /

Percentage

Male 12,500

Female 15, 500

Total 27,500

Sex Grade 11 Grade 12

F P F P

Male

Female

Total

- Populaton: Sex of all Grade 11 and 12 students in all public secondary schools in
Deped Leyte Division

: 2 Ways of Writing Sets

1. Rule Method

A = {xIx is the sex of a Grades 11 and 12 student, all public

secondary school in Deped Leyte Division}
 2. Roster Method

A = {F,F,F,F,F,F,F,M,M,M,M,M,...}

A = { 1,1,1,1,1,1,0,0,0,0,0,...}

*In statistics, population refers to values. Values of the variables.

Sample - is a subset of population (part of the population)

SAMPLING

•Process of getting samples from the population

SURVEY

•approach use in research

•Descriptive Research Design

•Uses the concept of sampling

¤Methods used in collecting the data

-Use of a survey questionnaire

-Use of an interview schedule

CENSUS

•Total enumeration

SAMPLING E

LO2: Discuss and cite examples of levels of measurement.

LO3: Clarify the two divisions of Statistics

LO4: Relate the concepts of population and sample to inferential statistics

LO5: Describe the different sampling techniques.

Lecture
February 24,2020

Statistical Hypothesis Testing

Hypothesis(ses)- educated (with a basis/ tentative but not absolute)

guess

Statistical Hypothesis-basically a statement that either negates or

declares the presence of either a significant difference of groups
with respect to a variable or a significant association between variables.

2 Types of Statisticsal Hypothesis

1. Null Hypothesis (H0)

-a statement negating the presence of either a significant difference of groups with respect
to a variable or a significant association between variables.

-examples

1. There is no significant difference in a achievement in Mathematics of students in public

and those in private schools.

2. There is no significant associations between academic performance in English and

Mathematics of students in public schools.

2. Alternative Hypothesis (H1)

-a statement declaring/affirming the presence of either a significant difference of groups

with respect to a variable or a significant association between variables.

Examples:

1. There is a significant difference in a achievement in Mathematics of students in public and

those in private schools.

2. There is a significant associations between academic performance in English and

Mathematics of students in public schools.

3 forms of alternative hypothesis:

1. One tailed test/unidirectional

2. One tailed test/unidirectional

3. Two tailed test/bidirectional

STEPS IN STATISTICAL HYPOTHESIS TESTING

STEP 1. Formulate the null and alternative hypothesis (HoHa)

STEP 2. Identify the level of significance (a) level of confidence

- Type I error.

A value (0.001, 0.01, 0.05, etc)

a= 0.01

- this means that if we repeat an experiment 100 times only 1 will have a different result.
Means 99 has the same result

a=0.05

this means that if we repeat an experiment 100 times only 5 will have a different result.
Means 95 has the same result

a=.001 his means that if we repeat an experiment 1000 times only 1 will have a
different result. Means 999 has the same result

Two types of error:

1. Type I error- error of deciding to reject a true null hypothesis (alpha)

2. Type II error-error on deciding not to reject a false null hypothesis (beta)

STEP 3. Decide the most appropriate test statistics and compute its value.

Step 4. Formulate the decision rule (D.R.)

Form of Ha Decision Rule (based on the Decision Rule (based on the

computed and tabular value p-value)
of the test statistic

Ha: > Reject Ho if tc > tsubscript a Reject Ho if the P-value is <

and v = ___, otgerwise fail to than alpha, otherwise fail to
reject Ho. reject Ho.

Ha: < Reject Ho -t subscript c < t

subscript a and v = -_____,
otherwise fail to reject Ho.

Ha: not equal Reject Ho if tc > tsubscript

a and v = ___, otgerwise Reject Ho if the p-value <a/2,
fail to reject Ho.
 Reject Ho -t subscript c < t otherwise fail to reject Ho.
subscript a and v = -_____,
otherwise fail to reject Ho.

Step 5. Formulate the decision based on the D.R. and the computed value of the test statistics.

Step 6. Formulate the conclusion based on the problem or scenario or purpose of the
reasearch.

Example:

The purpose of the research is to determine wether poverty incidence of municipalities relates
to the occurence of diarrhea outbreak in the municipality on the assumption that case
municipalities has higher poverty incidence. Use a 0.05 level of significance.

Step 1. HoHa

Ho: There is no significant difference between the poverty incidence and the case and
control munisiplities.

Ho: Mue zero = Mue one

Ha. There is a significant difference between the poverty incidence and the case and
control munisiplities,/ in fact case municipalities has higher poverty incidence.

Ha: Mue 0 < Mue 1

Ha: Mue is not equal to Mue 1

Step 2.

Alpha/ 2 = 0.05/2 = 0.025

Step 3.

T-test for independent samples

T= 2.896

P-value = 0.005

Step 4.
D.R. Reject Ho if the p-value <a/2 = 0.025, otherwise fail to reject Ho.

Step 5.

Decision: Since p= 0.005<a/2=0.025, REJECT Ho.So Ho is not true.

Step 6. Conclusion:

There is a significance difference in poverty incidence between the case and control
municipalities, in fact the case municipalities has significantly higher poverty incidence than
the control municipalities.Therefore, the occurence of the 2016 diarrhea outbreak in Eastern
Visayas significantly relates to poverty incidence of the municipalities.

Parametric-non-parametric-tests.pdf - Adobe Acrobat Reader DC

Parametric and Non Parametric test

In order to identify if you are going to use parametric and non parametric test you need to
identify if the data is normally distributed.

March 4, 2020

*Review of Elementary Statistics

-Basic Statistical Concepts (Universe, Variable, Popuation, sample)

-2 Division of Statistics

- Sampling (sample size, sampling techniques)

* Parametric vs Nonparametric tests

Parametric-normally distribution

Nonparametric-not equally distributed

* Steps in statistical hypothesis testing

- HoHa

- Alpha
 -Test Statistic (Spearman rho, Phi coefficient, Contingency coefficient, Point biserial
coefficient - Tests of Association)

Spearman rho - level of measure is interval or ordinal

- rank the values of data (transform)

- Used if data is inrerval or ratio but the data is not normally distributed

- D.R. Reject Ho if the p-value is less than or equal alpha, otherwise fail to reject Ho.

- Decision: Since the p-value... and alpha... decide. So, Ho is not true... Ho is true.

If you reject Ho then it is true.

If you fail to reject Ho then it is not true

- Conclusion: Therefore, there is a significant association between X and Y.

Chi Square

O = observed frequency

e = Expected frequency

Phi-coefficient (2 x 2 contingency table)

Contingency coefficient (5x5 contingency table)

D.R. Reject Ho if the computed Chi-square value > tabular Chi square value, itherwise
fail to reject

Decision: Since the computed Chi-square value= 18.596 > tabular Chi-square value=
9.488 at v=4, alpha. So Ho is not true.

Ho: There is no significant association between television viewing time and political
affiliation.

Ha: There is a significant association between television viewing time and political
affiliation.

Conclusion: There is a significant association between...

Hands-on experiencein using SPSS

 SPSS (Statistical Package for the Social Science)

From the data on community risk fators on the occurrence of diarrhea outbreak in 2016

Determine the ff.

1. Construct a frequency distribution table on the ff.

A. CaseCon

B. Climate Type

C. Wether a municipality has a sanitary waste disposal facility

D. Type of Sanitary Toilet

E. Number of Health Facilities

2. Descriptive Statistics (mean,standard deviation, maximum minimum) of the ff

variables.

Point-biserial (X (interval), Y (dichotomous))

Example:

Y= whether the municipality is a case or control municipality