Introduction to Statistics

Skezeer John B. Paz

Notre Dame of Marbel University

Statistics is a branch of
mathematics that deals with
the collection, organization,
presentation, analysis, and
interpretation of data.
Two Types of Statistics:

In each of the following, determine whether the Two Types of Data Gathering
given situation involves the use of DESCRIPTIVE 1. Census – methods of gathering data or
STATISTICS or INFERENTIAL STATISTICS. population wherein 100 percent of the
1. By 2040 at least 3.5 billion people will run total population is being asked.
short of water. -IS 2. Survey – method of gathering data or
2. Nine out of 10 on-the-job fatalities are men. population wherein only a representative
--DS sample of total population is being asked.
3. Expenditures for the cable industry were Population
$5.66 billion -it is the totality of all the objects of a certain
in 1996. class under consideration
-DS -it is a complete set of individual, objects or
4. The median household income for people measurements having some common
aged 25-34 is $35,888. observable characteristics
-DS Sample
- part of a population that has the same
5. Drinking decaffeinated coffee can raise characteristics of the given population
cholesterol Parameters
levels by 7%. -the value or measure obtained from the
-IS population
6. The national average annual medicine Estimates
expenditure - the value or measure obtained from the
per person is $1052. sample
DS Variables
7. Experts say that mortgage rates may soon hit - an observable characteristic or attribute
bottom. associated with the population or sample
IS being studied which makes one different from
the other. It can vary in quantity or in quality.
Introduction to Statistics
Skezeer John B. Paz
Notre Dame of Marbel University

Qualitative and Quantitative Variables between units of measure do exist; however,

Qualitative variables – are variables that there is no meaningful zero.
can be placed into distinct categories, Examples: SAT score, IQ, temperature
according to some characteristic or Ratio – possesses all the characteristics of
attribute interval
Examples: gender, religious preference, measurement, and there exists a true zero(a
geographic location point where none of the quality being measured
Quantitative variables – are numerical and exists). In addition, true ratio exist when the
can be ordered or ranked same variable is measured on two different
Examples: age, height, weight, body members of the population.
Temperature Examples: height, weight, time, salary, age

Classify each variable as Classify each as nominal, ordinal, interval or

qualitative or quantitative ratio level measurement.
1. Marital status of nurses in a hospital. 1. Groups of religion in Mindanao. nominal
Qualitative 2. Ranking of golfers in a tournament. ordinal
2. Time it takes to run a marathon. 3. Temperature inside 10 pizza ovens. interval
Quantitative 4. Weights of selected cell phones. ratio
3. Weights of lobsters in a tank in a restaurant. 5. Salaries of the coaches in the NFL. ratio
Quantitative 6. Times required to complete a chess game
4. Colors of automobiles in a shopping center ratio.
parking lot. 7. Ratings of textbooks. ordinal
Qualitative 8. Number of amps delivered by battery
5. Ounces of ice cream in a large milkshake. chargers. ratio
Quantitative 9. Ages of children in a day care center. ratio
6. Capacity of the NFL football stadiums. 10. Categories of magazines in a physicians
Quantitative office(sports, women’s, health, men’s, news)
7. Ages of people living in a personal care nominal
home.
Quantitative

Measurement scales Measures of Central

Nominal – classifies data into mutually exclusive Tendency and Variability
categories in which no order or ranking can be Prepared by:
imposed on the data. Skezeer John B. Paz
Examples: subject taught by college instructors, Notre Dame of Marbel University
sex, Measure of Central Tendency
political party, religion, marital status Ø gives a summary of the characteristics of a
Ordinal – classifies data into categories that can given set of data
beranked; however, precise differences between Ø mean, median, and mode
theranks do not exist. Mean
Examples: grade(A,B,C,D), judging(1st place, 2nd Ø one number that represents the whole data
place, etc.), rating scale (poor, good, excellent), Ø also known as the “average” of data
ranking of tennis players Median
Interval – ranks data and precise differences Ø middle number when the data are arranged
in order
Introduction to Statistics
Skezeer John B. Paz
Notre Dame of Marbel University

Mode
- score point with the greatest frequency

𝑵 - population size
Notations:

𝒏 - sample size
𝝁 - population mean
𝒙 - a specific score/value
𝒙 - sample mean
𝒙~ - sample median
𝒙 - sample mode
∑ - summation
Example #1:
Consider the following scores:
17, 14, 14, 21, 15, 16, 13, 24, 11

𝑥̅ = X1 + X2 +⋯+Xn/n =∑𝑥\𝑛
Sample Mean (𝒙)= 16.11

𝑥̅ =17+14+14+21+15+16+13+24+11/9
𝑥̅ = 145/9
𝑥̅ = 16.11
MEASURES OF VARIABILITY
Example #1:
Interpretation of Data
Consider the following scores:
Ø The lesser the value of the variability, the
17, 14, 14, 21, 15, 16, 13, 24, 11
more consistent, the more homogeneous and
𝜇 = X1 + X2 +⋯+Xn/n =∑𝑥\𝑛
Population Mean (𝝁)
the less scattered are the observations in the
𝜇 =17+14+14+21+15+16+13+24+11/9
set of data.
𝜇 =145/9
Ø If there is a large amount of variation, then
𝜇 =16.11
on average, the data values will be far from the
mean. Hence, SD will be large.
Ø If there is only a small amount of variation,
Example #1: then on average, the data values will be close
Consider the following scores: to the mean. Hence, SD will be small.
17, 14, 14, 21, 15, 16, 13, 24, 11
𝑹 - range
Notations:
Mean (𝒙)= 16.11
Median (~𝒙)= 15 𝝈 - population standard deviation
Mode (^𝒙) 14 𝒔 - sample standard deviation
𝝈2 - population variance
𝒔𝟐 - sample variance
Example #2:
92, 94, 94, 81, 85, 96, 93, 92, 91, 90

𝒙 =∑𝑥\𝑛
Mean=90.8

𝒙=908/10
𝒙 = 90.8 Mean=90.8
Introduction to Statistics
Skezeer John B. Paz
Notre Dame of Marbel University

Hypothesis
- is a conjecture about a population
parameter. This conjecture may or may
not be true.
Two Types of Hypothesis

- symbolized by 𝑯𝟎, is a statistical hypothesis

1. Null Hypothesis

that states that there is no difference

between a parameter and a specific value, or
that there is no difference between two
parameters.

- symbolized by 𝑯𝒂 (or 𝑯𝟏 ), is a statistical

2. Alternative Hypothesis

hypothesis that states the existence of a

difference between a parameter and a specific
value, or states that there is a difference
between two parameters.

Three Methods in Hypothesis Testing

2. The 𝑃-value method

1. The traditional method

3. The confidence interval method

1.Formulate the Null and Alternative Hypothesis

Null Hypothesis (𝐻!) - is the statement being
tested. If in case the null hypothesis is rejected,

Usually stated with equal sign: 𝑯𝟎: 𝝁𝟏 = 𝝁𝟐

then an alternative is to be accepted.

Alternative Hypothesis (𝐻% ) – possible values

about the population parameter.

a. 𝑯𝒂
Three ways of stating:

∶ 𝝁𝟏 ≠ 𝝁𝟐

b. 𝑯𝒂 ∶ 𝝁𝟏 > 𝝁𝟐 (one-tailed test)

(two-tailed test)

c. 𝑯𝒂 ∶ 𝝁𝟏 < 𝝁𝟐 (one-tailed test)

Types of Test Used with the Alternative

Hypothesis
a. One-Tailed Test
- Used when the rejection region is located at
only one extreme of the range value of the test
statistic, or it occupies only one side of the
normal curve.
- It is located only in one tail of the distribution
in the rejection region; either in the left tail or
the right tail ofthe distribution of the test
Introduction to Statistics
Skezeer John B. Paz
Notre Dame of Marbel University

statistic with an associated area of 𝛼 which also

𝐻o: 𝜇1 > 𝜇2
indicates a directional hypothesis.

𝐻a: 𝜇1 < 𝜇2

b. Two-Tailed Test
– a test which locates its critical region on both
tails of the distribution which also indicates a

𝐻a: 𝜇1 ≠ 𝜇2
Non-directional Hypothesis.

2. Set the Level of Significance (𝜶)

Terms:
Level of Confidence
- “belief”
- the degree of assurance that a particular
statistical statement is correct under specified

- represents the probability that 𝐻$ is true

conditions

Level of Significance 3. Determining the Statistical Test to be Used

- “doubt” A statistical test uses the data obtained from a
- the degree of uncertainty about the statistical sample to make a decision about whether the
statement under the same conditions used to null hypothesis should be rejected.

- represents the probability that 𝐻𝑜 is false

determine the confidence level The numerical value obtained from a statistical

The most commonly used values of 𝛼 are 0.01,

test is called the test value or test statistic.
3. Determining the Statistical Test to be Used
0.05 and 0.10. Choosing 0.01 level of If the test concerns means, the choices of the

a. 𝒛-Test
significance means that the researcher is 99% tests are:

If 𝜶 = 𝟎. 𝟎5
confident and has 1% to commit an error.
Basic Assumptions:
- the sample is a random sample

- either 𝑛 ≥ 30 or the population is normally

- population standard deviation (𝜎) is known

distributed if 𝑛 < 30
One-sample 𝒛 −Test
𝑧 =( ̅𝑥 − 𝜇) √ 𝑛/ 𝝈
b. 𝒕-test
Basic Assumptions:
- the sample is a random sample
- population standard deviation (𝜎) is unknown

- either 𝑛 ≥ 30 or the population is normally

(sample standard deviation (𝑠) is given instead)

distributed if 𝑛 < 30
One-sample 𝒕 −Test
𝑡 =( ̅𝑥 − 𝜇) √ 𝑛/s
4. Data Computation
Introduction to Statistics
Skezeer John B. Paz
Notre Dame of Marbel University

Ø Compute the value of the test statistic using

the formula.
5. Determining the Acceptance and Rejection
Regions
Ø The critical value separates the critical region
from the noncritical region.
Ø The critical / rejection region is the range of
values of the test value that indicates that there
is a significant difference and that the null 6. Compare the computed / test value and the
hypothesis should be rejected. critical / tabular value obtained
Ø The noncritical / nonrejection / acceptance Ø If the absolute computed value of the test
region isthe range of values of the test value statistic is less than the absolute critical /
that indicates that the difference was probably tabular value, accept the null hypothesis and
due to chance and that the null hypothesis reject the alternative hypothesis.
should not be rejected. Ø If the absolute computed value of the test
statistic is greater than or equal to the absolute
critical / tabularvalue, reject the null hypothesis
and accept the alternative hypothesis.
7. State your decision and conclusion.